Jack Christensen

Arrays are handles the same as any other type of parameter as far as the PostgreSQL server and the pgconn library are concerned. But pgconn is probably lower level than...

@zuckermanori Nothing has changed, but it looks look I made a mistake with the SQL. It should be `select * from table where value = any ($1)` instead.

Is there a particular reason for change? We only officially support Go 1.21+, but we don't purposely remove compatibility from older versions of Go either.

That would be convenient in many cases. But I'm concerned about how long the error message could be. `DETAIL` sometimes echos back the entire failing row. This could make the...

I'm not sure I understand the issue. The pooler sends an error and then closes the the connection? What level of error is it? I would expect it to be...

First, please ensure you are working on the correct repo. This repo is for the pgconn used with pgx v4. It was merged into the pgx repo for v5. As...

Have you considered using `CopyFrom`? Bulk inserts can have significantly higher performance through the copy protocol.

Actually, `CopyFrom` performs much better than you might think. From the pgx benchmarks (v5 but I think v4 is similar): ``` BenchmarkWrite2RowsViaInsert-12 9584 113651 ns/op 857 B/op 25 allocs/op BenchmarkWrite2RowsViaMultiInsert-12...

> It does appear that CopyFrom would still have the allocation issue I have above - https://github.com/jackc/pgx/blob/master/copy_from.go#L168C3-L168C3. Would it still be reasonable to consider a change to that write buffer?...

I think it is reasonable to make it easier to programmatically handle specific errors. But error constants and error types become part of the public interface and a tricky part...